[3.16] - March 31, 2022

This article provides an overview of new features in the Fortanix Confidential AI 3.16 release.

New Features

• This release adds support for "Bring Your Own Model". Users can now upload their own trained models (in ONNX or PMML format) and use them for inference. For more details refer to the following articles:
  • User’s Guide: Add Model
  • Example: Generate ONNX and PMML files for Logistic Regression using Bring You Own Model (BYOM)
• This release adds support for XGBoost algorithm that supports tabular datatypes and can be used to build a trained model on the datasets in Confidential AI. For more details refer to the following articles:
  • Confidential AI Models
  • Confidential AI User’s Guide
• This release also adds support for ResNet50 algorithm that supports image datatype for object recognition in Confidential AI. For more details refer to the article Confidential AI Models.

Limitations

Fortanix has a fair usage policy in this Early Customer Access Program. Hence, Fortanix has limited the resources one can create per account. Therefore, it is expected to observe a resource creation failure message once you have reached the max limit.

To report an issue/bug, visit https://support.fortanix.com/hc/en-us/requests/new.